Datasheet ATS612JSB Datasheet (Allegro)

Page 1
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
ATS612JSB
Data Sheet
27627.109
PRELIMINARY INFORMATION
(subject to change without notice)
November 20, 1998
1
2
3
4
Pin 1 = Supply Pin 2 = Output Pin 3 = Capacitor Pin 4 = Ground
Dwg. AH-006
DYNAMIC, SELF-CALIBRATING,
PEAK-DETECTING, DIFFERENTIAL
HALL-EFFECT GEAR-TOOTH SENSOR
The ATS612JSB dynamically-coupled gear tooth sensor is a peak detecting device that uses gain control to provide extremely accurate gear edge detection down to low operating speeds. Each sensor subassembly consists of a high-temperature plastic shell that holds together a samarium-cobalt magnet, a pole piece, and a dynamically­coupled differential open-collector Hall IC that has been optimized to the magnetic circuit. This small package can be easily assembled and used in conjunction with a wide variety of gear shapes and sizes.
The gear-sensing technology used for this sensor subassembly is Hall-effect based. The sensor incorporates a dual-element Hall IC that switches in response to differential magnetic signals created by ferrous targets. The sophisticated processing circuitry contains a self-calibrat­ing 5-bit A/D converter that normalizes the internal gain of the device to minimize the effect of air gap variations. The patented peak-detecting filter circuit eliminates magnet and system offsets and has the ability to discriminate relatively fast changes such as those caused by tilt, gear wobble, and eccentricities yet provides stable operation to extremely low RPMs.
ABSOLUTE MAXIMUM RATINGS
over operating temperature range
Supply Voltage, VCC...............................24 V*
Reverse Supply Voltage, V Output OFF Voltage, V Continuous Output Current,
............................ Internally Limited
I
OUT
Reverse Output Current, I Package Power Dissipation,
.......................................... See Graph
P
D
Operating Temperature Range,
................................. -40°C to +115°C*
T
A
Storage Temperature, TS.................. +170°C
* Operation at increased supply voltages with external circuitry is described in Applications Information. Devices for operation at increased temperatures are available on special order.
OUT
.............. -16 V
RCC
....................... 24 V
............ 50 mA
ROUT
These sensor systems are ideal for use in gathering speed, posi­tion, and timing information using gear-tooth-based configurations. The ATS612JSB is particularly suited to those applications that require extremely accurate duty cycle control or accurate edge detection similar to crank shaft applications. The lower vibration sensitivity also makes this device extremely useful for transmission speed sensing.
ATS612JSB: Large- or small-tooth gear-position sensing — crank angle, transmission speed, cam angle.
continued next page…
Always order by complete part number: ATS612JSB .
Page 2
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
FEATURES AND BENEFITS
Fully Optimized Differential Digital Gear-Tooth Sensor
Single-Chip Sensing IC for High Reliability
Digital Output Representing Target Profile
Extremely Low Timing Accuracy Drift with Temperature
Large Operating Air Gaps
Small Mechanical Size
Optimized Magnetic Circuit
Patented Peak-Detecting Filter:
80 µs Typical Power-On Time <10 RPM Operation (single-tooth target) Correct First-Edge Detection Uses Small Value Ceramic Capacitors
Under-Voltage Lockout
Wide Operating Voltage Range
Defined Power-On State
1000
800
R
θJA
600
400
200
0
ALLOWABLE PACKAGE POWER DISSIPATION IN mW
40 80 120
60 100 140 18020
AMBIENT TEMPERATURE IN °C
= 150°C/W
160
Dwg. GH-065-3
1
REG
MAGNET
Dwg. FH-014-1
SUPPLY
X
X
E1
E2
FUNCTIONAL BLOCK DIAGRAM
UVLO
GAIN
+ –
REFERENCE
POWER-ON
LOGIC
TRACK &
GENERATOR
HOLD
3
CAPACITOR
+ –
OUTPUT
2
CURRENT
LIMIT
GROUND
4
115 Northeast Cutoff, Box 15036 Worcester, Massachusetts 01615-0036 (508) 853-5000
W Copyright © 1998, Allegro MicroSystems, Inc.
Page 3
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
ELECTRICAL CHARACTERISTICS over operating voltage and temperature range, C3 = 0.1
Characteristic Symbol Test Conditions Min. Typ. Max. Units
µF to 0.47 µF.
Limits
Supply Voltage V Power-On State POS VCC = 0 5 V HIGH HIGH HIGH – Under-Voltage Lockout V Under-Voltage Hysteresis V Low Output Voltage V Output Current Limit I Output Leakage Current I Supply Current I
Power-On Delay t Output Rise Time t
Output Fall Time t
NOTE: Typical data is at VCC = 8 V and TA = +25°C and is for design information only.
CC
CC(UV) CC(hys)
OUT(SAT)
OUTM
OFF
CC
on
r
f
Operating, TJ < 165°C 3.6 24 V
I
= 20 mA, VCC = 0 5 V 2.5 3.6 V
OUT
Lockout (V I
OUT
V
OUT
V
OUT
Output OFF 6.0 8.2 12 mA Output ON 8.0 10 14 mA
VCC > 5 V 80 500 µs RL = 500 , CL = 10 pF 0.2 5.0 µs
RL = 500 , CL = 10 pF 0.2 5.0 µs
CC(UV)
= 20 mA 190 400 mV
= 12 V 25 45 55 mA = 24 V 0.2 15 µA
) – Shutdown 0.2 V
Page 4
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
OPERATION over operating voltage and temperature range with reference target (unless otherwise specified).
Limits
Characteristic Symbol Test Conditions Min. Typ. Max. Units
Air Gap Range AG Operating, 0.4 2.5 mm
Target Speed > 20 RPM
Calibration Cycle n
cal
Output edges before which 1 1 1 Edge calibration is completed*
Calibration Mode n
dis
Output falling edges for startup 64 64 64 Edges
Disable calibration to be complete Timing Accuracy t
θ
Target Speed = 1000 RPM, ±0.5 ±0.75 °
0.4 mm AG 2 mm
* Non-uniform magnetic profiles may require additional output pulses before calibration is completed.
REFERENCE TARGET
3 mm
t
h = 5 mm
TARGET
F (THICKNESS) 3 mm
D = 115 mm
o
T = 3 mm
E1 E2 SENSOR
POLE PIECE
SOUTH
PERMANENT
MAGNET
NORTH
1 2 3 4
AIR GAP
SIGNATURE
TOOTH (1)
Dwg. MH-016-2 mm
115 Northeast Cutoff, Box 15036 Worcester, Massachusetts 01615-0036 (508) 853-5000
Page 5
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
TYPICAL ELECTRICAL CHARACTERISTICS
14
12
11
10
9.0
8.0
7.0
SUPPLY CURRENT IN mA
6.0
4.0
-40
0 40 80 120
AMBIENT TEMPERATURE IN °C
OUTPUT ON
OUTPUT OFF
275
V = 8 V
CC
160
Dwg. GH-014-2
200
13
12
11
10
9.0
8.0
7.0
SUPPLY CURRENT IN mA
6.0
5.0
OUTPUT ON
OUTPUT OFF
T = 25°C
A
2.0 6.0 10 14 18
SUPPLY VOLTAGE IN VOLTS
Dwg. GH-058-1
250
225
200
175
I = 20 mA
OUT
OUTPUT SATURATION VOLTAGE IN mV
150
-40
0 40 80 120
AMBIENT TEMPERATURE IN °C
160
200
Dwg. GH-013-2
Page 6
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
TYPICAL OPERATING CHARACTERISTICS
(with reference target)
2.0
1000 RPM
TRAILING TARGET EDGE
1.6
1.2
0.8
0.4
RELATIVE ACCURACY IN DEGREES
0
-40°C
+25°C
+150°C
0.5 1.5 2.5
1.0 2.0 3.00
AIR GAP IN MILLIMETERS
2.0
1.6
SIGNATURE TOOTH
10 RPM
TRAILING TARGET EDGE
-40°C
+25°C
+150°C
Dwg. GH-008-7
2.0
1000 RPM
LEADING TARGET EDGE
1.6
1.2
0.8
0.4
-40°C
+25°C
+150°C
RELATIVE ACCURACY IN DEGREES
0
0.5 1.5 2.5
1.0 2.0 3.00
AIR GAP IN MILLIMETERS
2.0
1.6
SIGNATURE TOOTH
10 RPM
LEADING TARGET EDGE
-40°C
+25°C
+150°C
Dwg. GH-008-8
1.2
0.8
0.4
RELATIVE ACCURACY IN DEGREES
0
0.5 1.5 2.5
1.0 2.0 3.00
AIR GAP IN MILLIMETERS
1.2
0.8
0.4
RELATIVE ACCURACY IN DEGREES
0
Dwg. GH-008-6
0.5 1.5 2.5
1.0 2.0 3.00
AIR GAP IN MILLIMETERS
115 Northeast Cutoff, Box 15036 Worcester, Massachusetts 01615-0036 (508) 853-5000
Dwg. GH-008-5
Page 7
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
TYPICAL OPERATING CHARACTERISTICS
(with reference target) — Continued
1.0
0.8
0.6
0.4
0.2
PERIODIC TOOTH
1000 RPM
TRAILING TARGET EDGE
-40°C
+25°C
+150°C
RELATIVE ACCURACY IN DEGREES
0
0.5 1.5 2.5
1.0 2.0 3.00
AIR GAP IN MILLIMETERS
4.5
Dwg. GH-008-10
2.0
1.6
1.2
0.8
0.4
PERIODIC TOOTH
1000 RPM
LEADING TARGET EDGE
-40°C
+25°C
+150°C
RELATIVE ACCURACY IN DEGREES
0
0.5 1.5 2.5
1.0 2.0 3.00
AIR GAP IN MILLIMETERS
4.5
Dwg. GH-008-9
4.0
3.5
3.0
2.5
2.0
1.5
MAXIMUM AIR GAP IN MILLIMETERS
1.0 0
10 30 50 70
20 40 60
REFERENCE TARGET SPEED IN RPM
-40°C
+25°C
+150°C
Dwg. GH-011-4
4.0
3.5
3.0
2.5
2.0
1.5
MAXIMUM AIR GAP IN MILLIMETERS
1.0 0
500 1500 2500
1000 2000
REFERENCE TARGET SPEED IN RPM
-40°C
+25°C
+150°C
Dwg. GH-011-3
Page 8
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
DEVICE DESCRIPTION
The ATS612JSB dynamically coupled gear-tooth sensor system is a Hall IC/magnet configuration that is fully optimized to provide digital detection of gear-tooth edges in a small package size. This device contains self­calibrating circuitry that nulls out the effect of air gap variations on the switching accuracy of the device. A high startup hysteresis minimizes false switching caused by magnetic overshoot. The sensor is packaged in a minia­ture plastic housing that has been optimized for size, ease of assembly, and manufacturability. High operating temperature materials are used in all aspects of construc­tion.
The use of this sensor is simple; after correct power is applied to the component, it is capable of instantly providing digital information that is representative of the profile of a rotating gear. No additional optimization or processing circuitry is required. This ease of use should reduce design time and incremental assembly costs for most applications.
Sensing Technology. This gear-tooth sensor subassem­bly contains a single-chip differential Hall-effect sensor IC, a samarium-cobalt magnet, and a flat ferrous pole piece. The Hall IC consists of two Hall elements located so as to measure the magnetic gradient created by the passing of a ferrous object (a gear). The two elements measure the field gradient and convert it to voltage which is then subtracted and processed in order to provide a digital output signal.
The processing circuit uses a patented peak detec­tion scheme to eliminate magnet and system offsets. This technique allows dynamic coupling and filtering of offsets without the power-up and settling time disadvantages of classical high-pass filtering schemes. The peak signal of every tooth and valley is detected by the filter and is used to provide an instant reference for the operate and release point comparators. In this manner, the thresholds are adapted and referenced to individual signal peaks and valleys, hence providing immunity to zero line variation due to installation inaccuracies (tilt, rotation, and off center placement), as well as for variations caused by target and shaft eccentricities. The peak detection concept also allows extremely low speed operation for small value filter capacitors.
OPERATE
0
DIFFERENTIAL
MAGNETIC FLUX
V
BB
OUTPUT
V
OUT(SAT)
RELEASE
OPERATE
RELEASE
Dwg. WH-011
The ATS612JSB also includes self-calibration circuitry that is engaged at power on. The signal ampli­tude is measured and the device gain is normalized. In this manner, switch-point drift versus air gap is minimized and excellent timing accuracy can be achieved. The AGC circuitry, in conjunction with a unique hysteresis circuit, also eliminates the effect of gear edge overshoot as well as increases the immunity to false switching caused by gear tooth anomalies at close air gap. The AGC circuit sets the gain of the device after power on. Up to 0.25 mm air gap change can occur after calibration is complete without significant performance impact.
Superior Performance. The ATS612JSB peak-detecting differential gear-tooth sensor subassembly has several advantages over conventional Hall-effect gear-tooth sensors. The signal-processing techniques used in the ATS612JSB peak-detecting differential gear-tooth sensor solve the catastrophic issues that affect the functionality of conventional digital gear-tooth sensors.
Temperature drift. Changes in temperature do not
greatly affect this device due to the stable amplifier design and the offset rejection circuitry.
Timing accuracy variation due to air gap. The
accuracy variation caused by air gap changes is minimized by the self calibration circuitry. A 2x-to-3x improvement can be seen.
Dual edge detection. Because this device switches
from the positive and negative peaks of the signal, dual edge detection is guaranteed.
115 Northeast Cutoff, Box 15036 Worcester, Massachusetts 01615-0036 (508) 853-5000
Page 9
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
DEVICE DESCRIPTION — Continued
Tilted or off-center installation. Traditional differen­tial sensors will switch incorrectly due to baseline changes versus air gap caused by tilted or off center installation. The peak detector circuitry references the switch point from the peak and is immune to this failure mode. There may be a timing accuracy shift caused by this condition.
Large operating air gaps. Operating air gaps greater than 2.5 mm are easily achievable with this device due to the sensitive switch points after start up.
Immunity to magnetic overshoot. The air gap­dependent hysteresis minimizes the impact of over­shoot on the switching of device output.
Response to surface defects in the target. The gain­adjust circuitry reduces the effect of minor gear anoma­lies that would normally causes false switching.
Immunity to vibration and backlash. The gain-adjust circuitry keeps the hysteresis of the device roughly proportional to the peak to peak signal. This allows the device to have good immunity to vibration even when operating at close air gaps.
Immunity to gear run out. The differential sensor configuration eliminates the base line variations caused by gear run out.
-2000
-2500
-3000
-3500
-4000
-4500
SINGLE ELEMENT MAGNETIC FIELD IN GAUSS
-5000 0
10 20 30 60
ANGLE OF TARGET ROTATION IN DEGREES
Single-element flux maps
showing the impact of varying valley widths
TARGET
TARGET
T = 25°C
A
AIR GAP = 0.5 mm AIR GAP = 1.0 mm AIR GAP = 1.5 mm
AIR GAP = 2.0 mm
AIR GAP = 2.5 mm
5040
Dwg. GH-061-1
T = 25°C
A
Differential vs. Single-Element Sensing. The differential Hall-element configuration is superior in most applications to the classical single-element gear-tooth sensor. As shown in the flux maps on the next page, the single-element configuration commonly used (Hall-effect sensor mounted on the face of a simple permanent magnet) requires the detection of a small signal (often <100 G) that is superimposed on a large back-biased field, often 1500 G to 3500 G. For most gear/target configura­tions, the back-biased field values change due to concen­tration effects, resulting in a varying baseline with air gap, with valley widths, with eccentricities, and with vibration. The differential configuration cancels the effects of the back-biased field and avoids many of the issues presented by the single Hall element.
NOTE — 10 G = 1 mT, exactly.
1500
1000
500
0
-500
-1000
DIFFERENTIAL MAGNETIC FIELD IN GAUSS
-1500
0
AIR GAP = 0.5 mm
AIR GAP = 1.0 mm
AIR GAP = 2.5 mm AIR GAP = 2.0 mm AIR GAP = 1.5 mm
10 20 30 60
ANGLE OF TARGET ROTATION IN DEGREES
Differential flux maps vs. air gaps
5040
Dwg. GH-061
Page 10
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
DEVICE DESCRIPTION — Continued
Peak-Detecting vs. AC-Coupled Filters. High-pass filtering (normal ac coupling) is a commonly used tech­nique for eliminating circuit offsets. AC coupling has errors at power on because the filter circuit needs to hold the circuit zero value even though the circuit may power on over a large signal. Such filter techniques can only perform properly after the filter has been allowed to settle, which is typically greater than one second. Also, high-pass filter solutions cannot easily track rapidly changing baselines such as those caused by eccentrici­ties. Peak detection switches on the change in slope of the signal and is baseline independent at power up and during running.
Peak Detecting vs. Zero-Crossing Reference. The usual differential zero-crossing sensors are susceptible to false switching due to off-center and tilted installations, which result in a shift in baseline that changes with air gap. The track-and-hold peak-detection technique ignores baseline shifts versus air gaps and provides increased immunity to false switching. In addition, using track-and-hold peak-detecting techniques, increased air gap capabilities can be expected because a peak detector utilizes the entire peak-to-peak signal range as compared
to zero-crossing detectors that switch on one-half the peak-to-peak signal.
NOTE — “Baseline” refers to the zero-gauss differential where each Hall-effect element is subject to the same magnetic field strength.
Power-On Operation. The device will power on in the OFF state (output high) irrespective of the magnetic field condition. The power-up time of the circuit is no greater than 500 µs. The circuit is then ready to accurately detect the first target edge that results in a HIGH-to-LOW transi­tion.
Under-Voltage Lockout. When the supply voltage is below the minimum operating voltage (V device is OFF and stays OFF irrespective of the state of the magnetic field. This prevents false signals, which may be caused by under-voltage conditions (especially during turn on), from appearing at the output.
Output. The device output is an open-collector stage capable of sinking more than 20 mA. An external pull-up (resistor) to a supply voltage of not more than 24 V must be supplied.
CC(UV)
), the
115 Northeast Cutoff, Box 15036 Worcester, Massachusetts 01615-0036 (508) 853-5000
Page 11
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
CRITERIA FOR DEVICE QUALIFICATION
All Allegro sensors are subjected to stringent qualification requirements prior to being released to production. To
become qualified, except for the destructive ESD tests, no failures are permitted.
Test Method and Samples
Qualification Test Test Conditions Test Length Per Lot Comments
Temperature Humidity JESD22-A101, 1000 hrs 77 Device biased for Bias Life TA = 85°C, RH = 85% minimum power
Bias Life JESD22-A108, 1000 hrs 77
TA = 150°C, TJ = 165°C
(Surge Operating Life) JESD22-A108, 168 hrs 77
TA = 175°C, TJ = 190°C
Autoclave, Unbiased JESD22-A102, 96 hrs 77
TA = 121°C, 15 psig
High-Temperature JESD22-A103, 1000 hrs 77 (Bake) Storage Life TA = 170°C
Temperature Cycle JESD22-A104 1000 cycles 77 -55°C to +150°C ESD, CDF-AEC-Q100-002 Pre/Post 3 per Test to failure
Human Body Model Reading test Pin 3 > 1.5 kV
All other pins > 3 kV
GEAR/TARGET SYSTEM EVALUATION*
An analog map of the magnetic signal can be obtained by measuring the voltage at pin 3 (the capacitor pin) while the device is running. The peak and valley hold voltage will represent the peak-to-peak value of the signal. More accurate measurements can be taken by connecting an extremely small capacitor (0.05 µF) from pin 3 to ground. After the device is powered up and has switched at least 64 times, a 1 k bleed-off resistor should be installed in parallel with the capacitor without powering down the device. If the gear is then rotated at an ex­tremely low speed, an analog representation of the gain-
adjusted signal can be measured at pin 3. Note that the device should be re-powered at each air gap and the above procedure repeated for accurate measurements. In both cases, the analog signal may be compared to the typical hysteresis of the device and device performance can be estimated.
* In application, the terms “gear” and “target” are often interchanged. However, “gear” is preferred when motion is transferred.
Page 12
ATS612JSB
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
APPLICATIONS INFORMATION
Recommended Evaluation Technique. The self­calibrating feature of the ATS612JSB requires that a special evaluation technique be used to measure its high­accuracy performance capabilities. Installation inaccura­cies are calibrated out at power on only; hence, it is extremely important that the device be re-powered at each air gap when gathering timing accuracy data.
Self-Calibrating Functions. The ATS612JSB is designed to minimize performance variation caused by the large air gap variations resulting from installation by self-calibrating at power-on. This function should be tested using the following procedure.
1. Set the air gap to the desired value.
2. Power down and then power on the device.
3. Rotate the target at the desired speed.
4. Wait for calibration to complete (64 output pulses to
occur).
5. Monitor output for correct switching and measure
accuracy.
6. Repeat the above for multiple air gaps within the
operating range of the device.
7. This can be repeated over the entire operating tem-
perature range.
Measurement of the effect of changing air gap after power on:
1. Set the air gap to the desired value (nominal, for
example). Rotate the target at the desired speed. Apply power to the subassembly. Wait for 64 output pulses to occur. Monitor output for correct switching and measure accuracy.
Gear Diameter and Pitch. Signal frequency is a direct function of gear pitch and rotational speed (RPM). The width of the magnetic signal in degrees and, hence, the signal slope created by the tooth is directly proportional to the circumference of the gear (πDo). Smaller diameters limit the low-speed operation due to the slower rate of change of the magnetic signal per degree of gear rotation (here the limitation is the droop of the capacitor versus the signal change). Larger diameters limit high-speed opera­tion due to the higher rate of change of magnetic signal per degree of rotation (here the limitation is the maximum charge rate of the capacitor versus the rate of signal change). These devices are optimized for a 50 mm gear diameter (signal not limited by tooth width), 0.22 µF capacitor, and speeds of 10 RPM to 8000 RPM. For very large diameter gears (diameter >200 mm), the devices must be configured with a lower value capacitor, but not less than 0.1 µF. This allows for a range of 5:1 in gear diameters.
Air Gap and Tooth Geometry. Operating specifications are impacted by tooth width (T), valley width (pc - T) and depth (ht), gear material, and gear face thickness (F). The target can be a gear or a specially cut shaft-mounted tone wheel made of stamped ferrous metal. In general, the following gear or target guidelines must be followed to achieve greater than 2 mm air gap from the face of unit:
Tooth width, T.............................. >2 mm
Valley width, pc - T ...................... >2 mm
(Whole) depth, ht......................... >3 mm
Gear material............................... low-carbon steel
Gear face width (thickness), F..... >3 mm
Deviation from these guidelines will result in a
reduction of air gap and a deterioration in timing accuracy.
2. Change the air gap by ± 0.25 mm. Do not re-power
subassembly. Monitor the output for correct switching and measure accuracy.
115 Northeast Cutoff, Box 15036 Worcester, Massachusetts 01615-0036 (508) 853-5000
Page 13
ATS612JSB
A
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
APPLICATIONS INFORMATION — Continued
Operation with Fine-Pitch gears. For targets with a circular pitch of less than 4 mm, a performance improve­ment can be observed by rotating the front face of the sensor subassembly. This sensor rotation decreases the effective sensor-to-sensor spacing and increases the capability of detecting fine tooth or valley configurations, provided that the Hall elements are not rotated beyond the width of the target.
(mm)
(inch)
α
2.235 mm
0.088"
TARGET FACE WIDTH, F
>2.235 SIN
α
(mm)
>0.088 SIN
α
(inch)
α
α
2.235 COS
0.088 COS
A
Signal Timing Accuracy. The magnetic field profile width is defined by the sensor element spacing and narrows in degrees as the target diameter increases. This results in improved timing accuracy performance for larger gear diameters (for the same number of gear teeth). Valley-to-tooth transistions will generally provide better accuracy than tooth-to-valley transitions for large-tooth or large-valley configurations. For highest accuracy, targets greater than 100 mm in diameter should be used.
Signal Duty Cycle. For repetitive target structures, precise duty cycle is maintained over the operating air gap and temperature range due to an extremely good symme­try in the magnetic switch points and the internal self calibration of the device. For irregular tooth geometries, there will be a small but measureable change in pulse width versus air gap.
1 to pin 4), the output of the device will switch from LOW to HIGH as the leading edge of the target passes the subas­sembly, which means that the output will be HIGH when the unit is facing a tooth.
1
2
3
4
Dwg. AH-006-1
Dwg. MH-018-1
Power Supply Protection. The sensor contains an on­chip voltage regulator and can operate over a wide supply voltage range. For devices that need to operate from an unregulated power supply, transient protection should be added externally. For applications using a regulated supply, external EMI/RFI protection is often required. Insufficient protection can result in unexplained pulses on the output line, providing inaccurate sensing information to the user.
The filter capacitor and EMI protection circuitry can easily be added to a PC board for use with these devices. Provisions have been made for simple mounting of a board on the back of the unit.
Output Polarity. The output of the device will switch from HIGH to LOW as the leading edge of the target passes the subassembly in the direction indicated below (pin 4 to pin
1), which means that the output will be LOW when the unit is facing a tooth. If rotation is in the opposite direction (pin
4
3
2
1
Dwg. AH-007
Page 14
ATS612JSB
3
124
3
124
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
APPLICATIONS INFORMATION — Continued
Operation From a Regulated Power Supply. These devices require minimal protection circuitry during opera­tion from a low-voltage regulated line. The on-chip voltage regulator provides immunity to power supply variations between 3.6 V and 24 V. However, even while operating from a regulated line, some supply and output filtering is required to provide immunity to coupled and injected noise on the supply line. A basic RC low-pass circuit (R1C1) on the supply line and an optional output capacitor (C2) is recommended for operation in noisy environments. Because the device has an open-collector output, an output pull-up resistor (RL) must be included either at the sensor output (pin 2) or by the signal processor input.
SUPPLY
20
R
R
L
C
1
0.033 µF
1
Vcc
X
OUTPUT
0.22 µF
X
100 pF
C
3
C
2
+
-
Dwg. EH-008-1A
sufficiently high reverse breakdown capabilities so as to withstand the most negative transient. The current-limiting resistor (RZ) and the Zener diode should be sized for power dissipation requirements.
X
OUTPUT
0.22 µF
X
C
100 pF
C
3
2
+
-
Dwg. EH-008A
SUPPLY
2.5 k
0.033 µF C
R
L
20
R
1
R
Z
6.8 V
S
C
1
0.033 µF
Vcc
Capacitor Requirements. The choice of the capacitor at pin 3 (C3) defines the minimum operating speed of the target. This capacitor (0.1 µF minimum) is required to stabilize the internal amplifiers as well as to eliminate the signal offsets. Typically, a 0.22 µF low-leakage ceramic capacitor is recommended. Values greater than 0.47 µF should not be used as this may cause high-speed perfor­mance degradation.
Operation From an Unregulated Power Supply. In automotive applications, where the device receives its power from an unregulated supply such as the battery, full protection is generally required so that the device can withstand the many supply-side transients. Specifications for such transients vary between car manufacturers, and protection-circuit design should be optimized for each application. In the circuit below, a simple Zener-controlled regulator is constructed using discrete components. The RC low-pass filter on the supply line (R1C1) and a low-
dation in the low-speed performance of the device. Ex­cess capacitor leakage can result in the sensor changing output state without movement of the gear tooth being sensed. In addition to the capacitor leakage, it is ex­tremely important to minimize the leakage at the PC board and between the pins of the sensor. Up to 50 nA of external leakage can be tolerated at the capacitor pin node to ground. Choice of low-leakage-current potting com­pounds and the use of clean PC board techniques are extremely important.
value supply bypass capacitor (CS) can be included, if
Capacitor leakage current at pin 3 will cause degra-
necessary, so as to minimize susceptibility to EMI/RFI. The npn transistor should be chosen with sufficiently high forward breakdown voltage so as to withstand supply-side
Additional applications Information on gear-tooth
and other Hall-effect sensors is provided in the
Electronic Data Book
AMS-702 or
Application Note
Allegro
27701.
transients. The series diode should be chosen with
115 Northeast Cutoff, Box 15036 Worcester, Massachusetts 01615-0036 (508) 853-5000
Page 15
ATS612JSB
A
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
MECHANICAL INFORMATION
Component Material Function Units
Sensor Face Thermoset epoxy Maximum temperature 170°C* Plastic Housing Thermoplastic PBT, 264 psi deflection temp. (DTUL) 204°C
30% glass filled 66 psi deflection temp. (DTUL) 216°C
Approximate melting temperature 225°C Flame Class Rating UL94V-0 Leads Copper — Lead Finish 90/10 tin/lead solder plate † Lead Pull 8 N
* Temperature excursions to 225°C for 2 minutes or less are permitted. † All industry-accepted soldering techniques are permitted for these subassemblies provided the indicated maximum temperature for each component (e.g., sensor face, plastic housing) is not exceeded. Reasonable dwell times, which do not cause melting of the plastic housing, should be used.
Sensor Location (in millimeters)
(sensor location relative to package center is the design objective)
2.235
1.1
0.1
Dwg. MH-018 mm
Lead Cross Section (in millimeters)
0.41
0.0076 MIN. PLATING THICKNESS
0.38
Dwg. MH-019 mm
Page 16
ATS612JSB
A
DYNAMIC, SELF-CALIBRATING, PEAK-DETECTING, DIFFERENTIAL HALL-EFFECT GEAR-TOOTH SENSOR
DIMENSIONS IN MILLIMETERS
DIA
8.8
7.0
7.0
2.0
3.0
NOM
0.38
1.27
TYP
1 2 3 4
0.41
0.9
Tolerances, unless otherwise specified: 1 place ±0.1 mm, 2 places ±0.05 mm.
3.9
8.968.09
Dwg. MH-017-1B mm
Allegro MicroSystems, Inc. reserves the right to make, from time to time, such departures from the
detail specifications as may be required to permit improvements in the design of its products.
The information included herein is believed to be accurate and reliable. However, Allegro MicroSystems, Inc. assumes no responsibility for its use; nor for any infringements of patents or other rights of third parties which may result from its use.
115 Northeast Cutoff, Box 15036 Worcester, Massachusetts 01615-0036 (508) 853-5000
Loading...